Comparison of Strengthening Mechanism of the Nb, V, and Nb-V Micro-Alloyed High-Strength Bolt Steels Investigated by Microstructural Evolution and Strength Modeling

Abstract

The strengthening mechanism of Nb, V, and Nb-V micro-alloyed high-strength bolt steels was investigated and compared using microstructural evolution and strength modeling. Optical microscopy (OM), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and X-ray diffraction (XRD) were used to characterize the microstructure and precipitations. The results show that Nb-V composite micro-alloyed steel possessed a higher yield strength compared with Nb or V micro-alloyed steel when quenched at 870 °C and tempered at 450–650 °C. Furthermore, the strength increment of Nb-V micro-alloyed steel with respect to Nb or V micro-alloyed steel reached the maximum at a tempering temperature of 600 °C, and precipitation strengthening and dislocation strengthening presented higher strength contributions in Nb-V micro-alloyed steel than in Nb micro-alloyed steel and V micro-alloyed steel owing to the higher volume fraction and finer precipitate size. When V was added in combination with Nb in steel, the number of Nb-rich carbonitrides increased, which resulted in a higher volume fraction of the effective pinning particles-Nb-rich (Ti,Nb,V)(C,N) with diameters smaller than 50 nm and led to an enhanced refinement of the prior austenite grain. In addition, Nb could reduce the consumption of V during quenching, allowing more V to be solid-solved in the matrix after quenching, thereby further enhancing the precipitation strengthening effect of V during tempering.

1. Introduction

Fasteners have been widely used in the construction of machinery, automobiles, and long-span bridges [1,2,3,4,5]. In recent years, requirements for 10.9-grade high-strength bolts with a yield strength exceeding 940 MPa and an ultimate tensile strength exceeding 1040 MPa have been put forward for the increasing trends of high performance and lightweight of fastening parts [6,7]. Meanwhile, the risk of hydrogen-induced delayed fracture in bolts at such a high strength level may increase especially when applied in various harsh service environments [2,3,4,5]. Therefore, niobium (Nb), vanadium (V), and titanium (Ti) micro-alloying has been widely used in high-strength bolt martensitic steels, which is favorable for inhibiting steel softening without sacrificing toughness/hydrogen-induced cracking resistance during high-temperature tempering through the synergistic effect of mainly precipitation strengthening and grain refinement [2,5,8].

Micro-alloying elements like Nb, V, and Ti are often used independently or in combination with each other. Many studies have been conducted on their precipitation behavior during the heat treatment process and their effects on the grain refinement and strengthening mechanism. Titanium is commonly used to fix nitrogen, resulting in the formation of highly stable TiN particles within steel as it cools down from its solidified state, especially in steel containing boron (B) [9,10]. The presence of TiN particles inhibits the growth of austenite grains and promotes a finer microstructure upon transformation [11]. Additionally, Xu et al. [12] elucidated that the tensile strength of steels is positively correlated with the titanium content, with nano-sized Ti precipitates being the primary contributor to the enhanced strength of micro-alloyed steel. Meanwhile, it has been demonstrated in the research of Wang [11] that introducing Ti to steel with vanadium-bearing steel leads to a diminished strengthening contribution from V(C,N) precipitates, as the strengthening effect of V(C,N) precipitates is highly reliant on the nitrogen content available for the formation of V(C,N). It is thought that the reduction in the strengthening impact of V(C,N) following the addition of titanium is attributed to the decreased availability of nitrogen in the solution, which is necessary for the formation of TiN. However, Ti has no visible effect on Nb(C,N).

Wang [13] investigated the effect of Nb on the microstructure and the yield strength of 25CrMo47NbVTi martensitic steel and found that the microstructure was refined and the yield strength was enhanced remarkably with the addition of 0.030–0.060 wt.% Nb. In the temperature range from 700 to 1300 °C, both the precipitation temperature and the mass of Nb-rich (Nb,Ti,V)C were improved as the Nb content increased. Nevertheless, the precipitation behavior of Ti-rich (Ti,Nb,V)C was barely affected by the increasing Nb addition. The factors contributing to the increased yield strength can be linked to the sequential enhancement of grain boundary strengthening, precipitation hardening, and dislocation strengthening.

In the case of vanadium micro-alloyed steels, the precipitation of vanadium carbides typically occurs during the conventional tempering process, which is usually conducted within the temperature window of 600–700 °C [14,15,16]. Typically, when vanadium carbides precipitate within ferrite, they exhibit a Baker–Nutting orientation relationship with respect to the ferritic matrix, or vanadium carbides may form with a Kurdjumov–Sachs orientation in relation to the ferritic matrix when they precipitate from austenite [17]. Moreover, Lu [18] further claimed that MC-type carbides (where M represents V, Mo, or Cr) at the nano-scale have shown no significant coarsening even following a 10 h isothermal treatment at 600 °C, which is due to the incorporation of Mo/Cr atoms within the MC precipitates. This leads to particle size stabilization because of their higher chemical potentials compared with V, thereby contributing to a substantial precipitation strengthening of 236–382 MPa.

Zhang et al. [19] conducted a study to quantitatively compare the strengthening effects in three types of 20MnSi steel micro-alloyed with the elements of Nb, V, and Ti. They demonstrated that, among the alloying elements studied, Nb shows the most significant strengthening impact, whereas Ti exhibits the weakest strengthening influence. Moreover, the refined grain size and dispersed precipitations are the key contributors to steel strengthening in Nb or V micro-alloyed steel, whereas there is only precipitation strengthening in Ti-micro-alloyed steel. Nb-containing micro-alloyed steel has the finest average grain size among the four tested steels, whereas the grain size in Ti-containing micro-alloyed steel does not show significant refinement relative to the reference steel. Furthermore, the beneficial effects of multiple micro-alloying elements with respect to grain growth retardation or precipitation behaviors during the soaking process have also been established. For example, the evolution of austenite grain in steels containing Nb-Ti was compared with that of austenite grain in steels containing Nb-V by Karmakar [20]. They found that the smallest possible austenite grain size in Nb-V steel when soaked at a lower temperature (<1075 °C) is due to the grain boundary pinning effect caused by AlN, Nb(C,N), and V(C,N) precipitates. At temperatures of 1150–1200 °C, the grain size bimodality appeared and was more distinct in Nb-V steel. The addition of Ti can reduce the grain size bimodality in Nb steels but cannot prevent it. Meanwhile, the higher stability of TiN precipitates inhibits grain growth in Nb-Ti steel during soaking at elevated temperatures.

Nevertheless, there has been limited investigation on the essential differences between the influence of composite micro-alloying and that of single micro-alloying on the microstructure, specifically in terms of precipitate size and volume fraction and grain size, as well as the final mechanical properties. In the current work, the relation between the microstructural evolution and strengthening mechanisms in steels micro-alloyed with Nb, V, and composite Nb-V has been systematically investigated and compared by multi-scale characterizations and modeling. This will facilitate the development of a straightforward and cost-efficient approach for producing high-strength bolt steel with comprehensive, excellent properties.

2. Materials and Methods

Three different micro-alloyed steels were produced in the lab through vacuum induction melting and subsequently formed into ingots weighing approximately 100 kg. The actual chemical compositions of all three steels are given in

Table 1. Chemical compositions of the steels (wt.%).

Steel	C	Mn	Si	S	P	Mo	Cr	V	Ti	Nb	B	N
Nb-V	0.278	0.7	0.25	≤0.01	≤0.01	0.2	0.71	0.05	0.015	0.03	0.0015	0.0034
Nb	0.282	0.7	0.24	≤0.01	≤0.01	0.2	0.74	-	0.015	0.03	0.0015	0.0040
V	0.277	0.7	0.25	≤0.01	≤0.01	0.2	0.70	0.05	0.015	-	0.0015	0.0032

, labeled as Nb-V steel, Nb steel, and V steel. The cylindrical billet underwent a soaking process at 1200 °C for 3 h, and then it was subjected to hot rolling at temperatures ranging from 1100 °C to 1150 °C to achieve a thickness of 20 mm. Subsequent to rolling, the plates were rapidly cooled in water down to 700 °C before being allowed to cool to ambient temperature in air.

The 16 × 120 × 180 mm³ plates underwent the same heat treatment regimen, which consisted of austenitizing at 875 °C for a period of 60 min, were subsequently quenched in water, and finally were tempered in the range of 450–650 °C for an additional 60 min.

Tensile test samples conforming to the ASTM A370-09 specification [21] were cut from the tempered plates following a longitudinal orientation. The tensile testing was performed on a CMT5305 Instron (Metus Industrial Systems Co., Ltd., Eden Prairie, MN, USA) apparatus at ambient temperature.

The as-quenched specimens were subjected to etching in a supersaturated picric acid water solution to facilitate optical microscopy (OM) (Axiover, Germany) examination of the prior austenite grain (PAG).

The samples in the as-tempered condition were etched in 4% nital for martensitic packet observations through a Hitachi S-3400N scanning electron microscope (SEM) (Hitachi, Chiyoda, Japan). The martensitic block and the misorientation angles between different grain boundaries were disclosed by mechanical polishing with silica suspension, and electron backscattered diffraction (EBSD) (TexSemLaboratories, Inc., San Diego, CA, USA) maps were acquired by a field emission microscope furnished with a TSL orientation imaging microscopy (OIM) (5.0, Oxford Instruments Company, London, UK) system. Thin foils and carbon extraction replicas were used to characterize the dislocations, martensitic lath, and precipitates through a JEM-2010 HR-TEM (JEOL, Tokyo, Japan).

The sizes of the PAG, packet, and block were statistically evaluated by averaging at least 100 grains from the corresponding micrographs. In order to determine the average equivalent size and volume percentage of precipitates in different shapes, all of them are treated in the same way as spherical particles. The equivalent diameter (d) is taken as the average particle size. Then, the volume fraction of the particles f is calculated according to the following equation:

$$f={\displaystyle \frac{N\frac{4\pi }{3}}{S_{0}d}}{\left({\displaystyle \frac{d}{2}}\right)}^3={\displaystyle \frac{N\pi d^2}{6S_0}}={\displaystyle \frac{2NS}{3S_0}}$$

(1)

Here, N represents the number of particles per area. S₀ is the specific area for estimation, and S stands for the particle area.

X-ray diffraction (XRD) analysis was used to determine the dislocation density of the sample. The XRD spectra were obtained by scanning using a Rigaku D/max-2500/PC diffractometer (Rigaku, Tokyo, Japan). The scanning angle (2θ) ranged from 30° to 120°, and the step size was 0.02°. The full width half maximum values were computed by using a modified Williamson–Hall method to decouple the microstrain and grain size broadening effects [22]. The microstrain term determined from the modified Williamson–Hall [23,24] method was regarded as being mainly related to the dislocation density.

3. Results

3.1. Yield Strength (YS)

The room-temperature tensile properties of steels with different Nb and V contents tempered at different temperatures were measured. An engineering stress–strain curve diagram was provided, and the YS was determined by the 0.2% offset flow stress (R_t0.2), as presented in Figure 1. It demonstrates that the YS decreased from 1271 MPa, 1253 MPa, and 1221 MPa to 920 MPa, 915 MPa, and 860 MPa, respectively, along with a tempering temperature increase from 450 °C to 650 °C for Nb-V, Nb, and V steel, suggesting a marked decrease in YS or a gradual softening behavior of lath martensite, as a result of tempering at an elevated temperature. Nb-V composite micro-alloyed steel possessed the highest yield strength compared with Nb or V micro-alloyed steel when quenched at 870 °C and tempered at 450–650 °C. Furthermore, the strength increment of Nb-V micro-alloyed steel with respect to Nb or V micro-alloyed steel reached the maximum at a tempering temperature of 600 °C. Therefore, the samples quenched at 875 °C and tempered at 600 °C were selected for further microstructure and precipitation analysis.

3.2. Microstructural Observations

3.2.1. PAG Observations

Figure 2a–c illustrate the characteristic prior austenite grain (PAG) shapes in as-quenched specimens containing different Nb and V contents. The figures reveal that equiaxed PAGs distributed homogeneously in each steel. The average PAG size, D_c, for each steel was quantitatively measured and is presented in

Table 2. The quantitative microstructural parameters used for strength modeling.

Steel/MPa	Dc/µm	Dp/µm	Db/µm	ρdis/1015 m−2	dpre/nm	fpre/%
Nb-V	16.5	9.1	1.71	1.15	15.5	6.4
V	23.9	13.2	1.79	1.00	16.9	5.9
Nb	20.4	11.5	1.92	0.96	19.7	5.2

D_c—average grain size of the PAG; D_p—mean size of the packet; D_b—mean size of the block; ρ_dis—density of the dislocations; d_pre—mean equivalent diameter of the precipitates; f_pre—volume fraction of the precipitate.

. It indicates that 0.03 wt.% Nb has a more significant refining effect on the prior austenite grain than 0.05 wt.% V. Furthermore, with the addition of 0.03 wt.% Nb and 0.05 wt.% V composite into steel, the prior austenite grain of steel can be further refined to 16.5 μm.

3.2.2. Packet and Block Observations

It is well known that a martensite structure is a hierarchical structure and that several martensite laths form one block. One of these blocks with the same orientation is part of a packet, and a prior austenite grain (PAG) can be divided into packets that have the same habit plane [25,26]. In order to study the effect of Nb/V on a martensitic structure in greater detail, the characteristic shapes of martensite packets and blocks within the tempered specimens were examined using scanning electron microscopy (SEM) and electron backscatter diffraction (EBSD), as shown in Figure 3 and Figure 4, respectively. The packet size, D_p, and block size, D_b, were quantitatively determined, as shown in Table 2. It indicated that the tendency of D_p and D_b was essentially consistent with that of D_c, which was consistent with our previous results.

3.3. Precipitation Observations

3.3.1. Precipitation of As-Quenched Samples

As suggested by Yu et al. [27], the PAG boundaries can be pinned and further refined by the precipitates of a certain diameter and quantity. Therefore, the precipitates in as-quenched samples with varied Nb and V contents were observed using TEM and are shown in Figure 5a–f, and the corresponding EDX results of Type 1–Type 6 particles are shown in Figure 5g–l. Two types of precipitates were found in Nb-V composite micro-alloying steels: rectangle Ti-rich(Ti,Nb)C/50–130 nm particles (a small number of above particles contain the element N) for Type 1 and elliptic Nb-rich(Ti,Nb,V)C/5–50 nm particles for Type 2. The Type 1 and Type 2 particles are marked by blue arrows and a red ellipse, respectively, in Figure 5b, and the corresponding EDX results are shown in Figure 5g and Figure 5h, respectively. For V steel, two types of precipitates were detected: Ti(C,N)/50–160 nm particle for Type 3 and Ti-rich(Ti,V) (C,N)/40–140 nm particle for Type 4. The Type 3 and Type 4 particles are marked by yellow and green arrows, respectively, in Figure 5d, and the corresponding EDX results are shown in Figure 5i and Figure 5j, respectively. For Nb steel, another two types of precipitates were detected: elliptic and irregular Ti-rich(Ti,Nb)(C,N)/30–120 nm particle for Type 5 and Nb-rich(Ti,Nb)(C,N)//5–15 nm particle with an elliptic shape for Type 6. Moreover, the irregular Nb-rich(Ti,Nb)(C,N) is presumed to be composed of undissolved Ti(C,N) carbonitride and newly precipitated (Nb,Ti) (C,N) carbide according to the work of Jung [28]. The Type 5 and Type 6 particles are marked by black arrows and a white ellipse, respectively, in Figure 5f, and the corresponding EDX results are shown in Figure 5k and Figure 5l, respectively.

Several studies [9,29] proposed that the average sizes and volume percentage of second-phase particles during quenching affect the pinning force, which is crucial in hindering the growth of austenite grains. Therefore, the diameter distribution and the volume fractions during the different size intervals for as-quenched steels with different Nb and V contents are summarized in Figure 6 and Figure 7, respectively. As indicated by Arribas [30], precipitates smaller than 50 nm have strong pinning effects on austenite grain boundaries, and as the size of the precipitates increases, their pinning effect weakens. Figure 6 shows that, for Nb-V steel, about 90% of the precipitates are smaller than 50 nm. Combined with TEM observation results, it can be seen that these fine precipitates are mainly contributed by Nb-rich (Ti,Nb,V)C and some Ti-rich (Ti,Nb)C particles. For V steel, only about 65% of the precipitates are smaller than 50 nm. As observed using TEM, these fine precipitates are primarily attributed to the smaller-sized Ti-rich (Ti,V)(C,N) particles. For Nb steel, about 85% of the precipitates are smaller than 50 nm. According to TEM observations, these fine precipitates are mainly contributed by Nb-rich (Ti,Nb)(C,V) particles. In addition, as can be seen in Figure 7, the volume percentage of small-sized particles shows a roughly decreasing trend in the order of Nb-V, Nb, and V steel, while the large-sized particles exhibit an increasing trend.

3.3.2. Precipitation of As-Tempered Samples

Figure 8a–c show transmission electron microscopy (TEM) images of the tempered samples containing different Nb and V concentrations. The martensite in each of Nb-V steel, Nb steel, and V steel is in a regular lath shape with a width of about 0.9 μm. Diverse types of precipitates, varying in both shape and size, were observed within the matrix and at the lath boundaries. The average equivalent diameter d_pre and volume fraction f_pre of the precipitates for as-tempered samples with various Nb and V contents are quantified and shown in Table 2. It can be seen in Table 2 that the average equivalent diameter d_pre of the precipitates exhibits a trend of gradual decrease in the order of Nb-V, V, and Nb micro-alloyed steel, while the volume percentage f_pre of the precipitates shows a decreasing trend.

3.4. Dislocation Observations

Figure 8 illustrates the dislocation morphology of steels containing varying Nb and V contents in the as-tempered conditions. The observations also reveal that both steels exhibit entangled dislocation networks clustering around the precipitates. More dislocations were pinned in Nb-V steel compared with V steel and Nb steel and thus might effectively improve the YS [31]. The XRD pattern of each as-tempered steel is presented in Figure 9, and the dislocation density of steels with different Nb and V contents estimated in Figure 9 is shown in Table 2. The dislocation density of Nb-V steel is 1.15 × 10¹⁵ m⁻² and is higher than that of V steel and Nb steel, the dislocation densities of which are 1.00 × 10¹⁵ m⁻² and 0.96 × 10¹⁵ m⁻², respectively. On one hand, according to Abe [32], subdrain boundaries can function as barriers to dislocation motion. Thus, it can be deduced that the enhanced refinement of the martensite microstructure due to the addition of composite micro-alloying Nb and V may contribute to the increase in dislocations. On the other hand, the precipitation produced by quenching and tempering can also pin dislocations, preventing them from annihilating. By comparing V steel and Nb steel, it can be inferred that, during the tempering process, precipitates with an excellent strong pinning effect may have been produced in V steel, resulting in a higher dislocation density than that of Nb steel.

4. Discussion

4.1. Effect of Nb and V Micro-Alloying on the Precipitates in As-Quenched Specimens

For a better understanding of equilibrium phases during quenching and tempering for grain size control, thermodynamic calculations, using Thermo-Calc with the database TCFE7, were used. Thermodynamic calculations were performed with full compositions for Nb-V steel, V steel, and Nb steel for identifying possible equilibrium phases. An equilibrium phase diagram of each steel is shown in Figure 10.

It can be seen in Figure 10a that, for Nb-V steel, the equilibrium phases are FCC_A1#2, FCC_A1#3, cementite, M₂₃C₆, MC, and M₇C₃ according to the precipitation temperature from high to low in the whole temperature range from 1600 °C to 400 °C. Moreover, there are two kinds of particles above the quenching temperature of 875 °C, namely, FCC_A1#2 and FCC_A1#3. By analyzing the compositions and their contents in each phase, it is apparent that FCC_A1#2 is Ti-rich (Ti,Nb,V)(C,N), corresponding to the Type 1 particle (Ti-rich (Ti,Nb)C/50–130 nm) observed in as-quenched samples of Nb-V steel. FCC_A1#3 is Nb-rich (Ti,Nb)C, corresponding to the Type 2 particle (Nb-rich (Ti,Nb,V)C/5–50 nm). Although there is a slight difference in composition between the calculated and the TEM observation results, this is considered inevitable and has no effect on the analysis results.

The precipitation temperature and mass percentage of the above two phases at the quenching and tempering temperatures can be obtained from Figure 10 and are shown in

Table 3. The precipitation temperature and the mass percentage at the quenching and tempering temperatures.

	Precipitate Type	Mass Fraction/wt.% at 600 °C	Mass Fraction/wt.% at 875 °C	Precipitation Temperature/°C
V	MC (rich V, Mo, Ti, C)	0.0013812	-	750
	FCCA1#2 (rich Ti, N, C)	0.0001916	0.0001980	1400
Nb	FCCA1#2 (rich Nb, Ti, C, N)	0.0005168	0.0003606	1220
	FCCA1#3 (rich Ti, Nb, N, C)	0.0001774	0.0001606	1400
Nb + V	MC (rich V, Mo, Ti, C)	0.0014015	-	740
	FCCA1#3 (rich Nb, Ti, C, V)	0.0005496	0.0003929	1220
	FCCA1#2 (rich Ti, Nb, V, C, N)	0.0001728	0.0001522	1400

. It can be seen that the precipitation temperature of Type 1 particles is 1400 °C, and its mass percentage is 0.01522 wt.% under a quenching temperature of 875 °C. The precipitation temperature of Type 2 particle is 1220 °C, and its mass percentage under the as-quenched state is 0.03929 wt.%. As the precipitation temperatures of Type 1 and Type 2 particles were both higher than 875 °C, they were likely to precipitate in the rolling and subsequent cooling process. Moreover, the lower driving force for precipitation growth due to the lower precipitate supersaturation during the rolling process might lead to smaller sizes of the Type 2 particles than that of Type 1. When steel was reheated at 875 °C, the large and stable Type 1 particles tended to continue growing, while the small Type 2 particles were likely to dissolve, and re-precipitation occurred as a result of the reduced solubility in austenite, and the N–W orientation between these particles and austenite restricted the growth, resulting in a relatively small size (10–50 nm) of the particles in the as-quenched specimens [33]. In addition, the higher precipitated mass percentage and smaller size may explain the higher numbers of Type 2 particles.

For V steel, there was only one type of particle above the quenching temperature of 875 °C, namely, FCC_A1#2. By analyzing the composition and the contents in the phase, FCC_A1#2 is (Ti,V)(C,N) corresponding to the Type 3 particle (Ti-rich Ti(C,N)/50–160 nm) and Type 4 particle (Ti,V)(C,N)/40–140 nm observed in as-quenched samples. The precipitation temperature of Type 5 and Type 6 particles is 1400 °C, and the mass percentage of the precipitated phase is 0.01980 wt.% in the as-quenched state. Similar to Type 1 particles, they were formed during the rolling process and continued to grow during the reheating process, resulting in particles of larger sizes.

For Nb steel, the precipitates of FCC_A1#3 and FCC_A1#2 existed above the quenching temperature of 875 °C. It can be seen by examining the composition and their contents in each phase that FCC_A1#3 is Ti-rich (Ti,Nb)(C,N), corresponding to the Type 6 particle (Ti-rich (Ti,Nb)(C,N)/30–120 nm) observed in the as-quenched samples. FCC_A1#2 is Nb-rich (Ti,Nb)(C,N), corresponding to the Type 5 particle. As indicated in Table 3, the precipitation temperature of a Type 5 particle is 1400 °C, which is similar to that of Type 1 particles in Nb-V steel, and its mass percentage in the as-quenched state is 0.01606 wt.%. The precipitation temperature of the Type 5 particle is 1220 °C, which is similar to that of the Type 2 particles in Nb-V steel, and its mass percentage reaches 0.03606 wt.% in the as-quenched state.

According to the above analysis, it can be seen that Ti-rich complex precipitates that have large sizes are all present in each of the three types of steel. Compared with Nb steel and V steel, the combined addition of Nb and V has little effect on the precipitation temperature of Ti-rich complex precipitates, but its mass percentage is reduced. Since the Ti-rich complex precipitates usually incorporate the elements of Nb and V, the decrease in the mass percentage of Ti-rich complex precipitates can allow more V and Nb to be dissolved in the matrix, which may be used to form small dispersed MC precipitation as much as possible during the tempering process and have a stronger strengthening effect on steel.

As for the Nb-rich complex precipitates in Nb steel and Nb-V steel, it is known that the relatively low driving force for precipitation growth may account for the small sizes of the particles. Compared with Nb steel, the combined addition of Nb and V in Nb-V steel results in an increase in precipitation mass percentage. Therefore, the Nb-containing precipitates in Nb-V steel has a smaller size and a higher amount and further results in the smallest average equivalent diameter and the highest volume fraction of the particles smaller than 50 nm in Nb-V steel under the as-quenched state.

Meanwhile, due to the low precipitation temperature of V-rich precipitates, there are very few new V-rich precipitates formed during isothermal treatment at 875 °C. Therefore, the small number of large-sized Ti-rich precipitates and fine Ti-rich (Ti,V)(C,N) precipitates that were formed during the rolling process leads to the largest equivalent diameter and the smallest volume fraction of the particles smaller than 50 nm in V steel under the as-quenched state.

4.2. Effect of Nb and V Micro-Alloying on the Precipitates in As-Tempered Specimens

When steels were tempered at 600 °C, precipitates such as M₃C, M₂₃C₆, and M₇C₃, with Fe, Cr, and Mo as the main elements, began to precipitate, and these precipitates were generally distributed along the lath boundaries, which generally provide fast diffusion channels for atoms to help precipitates to nucleate and growth. Then, these precipitates coarsened to a relatively large size and gave limited yield strength contributions. Therefore, we will not discuss them further in detail here. In the meanwhile, the Type 1–Type 6 particles formed during the quenching process continued to grow or newly precipitate due to the decreased solubility of atoms in ferrite. However, the mass percentage of these precipitates was very little according to thermodynamic calculation results. Additionally, we found that, for steels containing V, e.g., Nb-V steel and V steel, V-rich MC particles began to precipitate, as shown in Figure 11 and Figure 12. These particles are small, with only 5–10 nm in size, and have a small lattice mismatch with the matrix, indicating a coherent or semi-coherent relation to the matrix, and usually have a strong pinning effect on dislocations, which is very helpful for improving the strength of steel [18]. Moreover, the thermodynamic calculation results in Table 3 show that the mass percentage of MC precipitates in Nb-V steel is more than that in V steel. This may be explained that, for Nb-V steel, V only participated in the formation of Type 2 particles (Nb-rich (Ti,Nb,V)C/5–50 nm) during the quenching process. Since the V content in Type 2 particles is low and the particle size is small, most of the V remains dissolved in the matrix and could participate in the precipitation of MC during the subsequent tempering process. In contrast, for V steel, V participated in the formation of Type 6 particles (Ti-rich (Ti,V)(C,N)/40–140 nm), which had a relatively larger size due to the high precipitation temperature. There is relatively less V atom solid solution in the matrix after quenching to participate in the formation of MC. Finally, fewer MC particles formed in V steel than that in Nb-V steel. In other words, the combined addition of Nb-V could further improve the precipitation strengthening effect of V.

Additionally, nano-sized precipitates were also found in Nb steel, as shown in Figure 13. High-resolution images combined with energy spectrum analysis reveal that these precipitates are NbC particles. However, due to the consumption of Nb by a large number of Type 1 particles (Ti-rich (Ti,Nb)C/50–130 nm) and Type 2 particles (Nb-rich (Ti,Nb,V)C/5–50 nm) during the quenching process, only a small amount of Nb remains dissolved in the matrix that could be used in the formation of the NbC particles during the subsequent tempering process. Therefore, the Nb-containing precipitates found in the tempered specimens were in fact formed in the quenching process. These particles might grow in the subsequent tempering process, leading to a relatively larger size and lower precipitation strengthening contributions to Nb steel.

4.3. Effect of Precipitates on Austenite Grain and Martensitic Structure Refinement

The method for reducing austenite grain size during reheating is known to be governed by the phase transformation and pinning effect by solute atoms or precipitated particles. Once the transformation to austenite is finalized, the size of the austenite grains is predominantly controlled by the latter. The solved alloying elements in austenite and the fine and dispersed carbonitride precipitates can impede the growth of austenite grains via solute drag [27] and Zener drag [34], respectively. Meanwhile, the growth of austenite grains is mainly controlled by the Zener drag at relatively lower temperatures, under which carbonitride might precipitate due to the decreased solubility of dissolved atoms, and the solute drag is usually negligible [20]. The grain boundaries curve around the second-phase particles during the growth process. Furthermore, the following formula shows the quantitative relationship between austenite grain size and the characteristic parameter of the precipitates [35,36], as follows:

$$D_{\mathrm{c}}={\displaystyle \frac{\pi d_{\mathrm{pre}\text{-}\mathrm{q}}}{6f_{\mathrm{pre}\text{-}\mathrm{q}}}}\left({\displaystyle \frac{3}{2}}-{\displaystyle \frac{2}{Z}}\right)$$

(2)

where D_c represents the austenite grain size, d_pre-q and f_pre-q indicate the average diameter and the volume fraction of the precipitates during the austenitizing process, and Z denotes the ratio of the radii of growing grains and matrix grains; the value of Z is generally assumed to be a constant ranging from 1.5 to 2.0 [20,37]. Some studies [30] reported that particles smaller than 50 nm were effective in controlling austenite grain growth. Therefore, the volume fraction f_pre-q and the average diameter d_pre-q of the precipitate smaller than 50 nm were quantified. f_pre-q was 0.058%, 0.051%, and 0.047%, and d_pre-q was 12.7 nm, 13.3 nm, and 16.4 nm for as-quenched Nb-V steel, Nb steel, and V steel, respectively. Based on the results, the austenite grain sizes for each steel are calculated using Equation (2) and presented in Figure 14. It is found that the predicted austenite grain sizes are consistent with the experimental results, indicating that Type 2 (Nb-rich (Ti,Nb,V)C/5–50 nm) formed in Nb-V steel and Type 3 (Nb-rich (Ti,Nb)(C,V)/30–120 nm) and Type4 (NbC/5–15 nm) nanoprecipitation formed in Nb steel are very effective in retarding austenite grain growth. Furthermore, the observed value is lower than the predicted value; the reason for the difference is the weak pinning effect of precipitates larger than 50 nm on the grain boundaries. Our prior studies have shown that the packet and block structures within martensite can be concurrently refined through PAG refinement [9]; the sizes of the packet and block in the Nb-V steel and Nb steel were also smaller than that of V steel.

4.4. Quantification of the Contribution from Various Hardening Factors

To examine the impact of microstructure on yield strength, the distinct microstructural factors influencing yield strength are simulated. The lattice friction strengthening σ₀ is set to 50 MPa [38].

The grain refinement strengthening σ_gb depends on the estimated grain size in steel and can be calculated using the following Hall–Petch equation [39]:

$${\sigma }_{\mathrm{gb}}=K_{\mathrm{HP}}{D_{\mathrm{b}}}^{\raisebox{1ex}{$-1$}\!\left/ \!\raisebox{-1ex}{$2$}\right.}$$

(3)

where K_HP = 0.21 MPa m^1/2 is the Hall–Petch constant for martensite [16], and D_b is the block size, which was considered as the strength control unit measured based on the EBSD technique.

The dislocation strengthening σ_dis provides a major contribution to the high strength of the steel and can be calculated using the following Bailey–Hirsch relationship [22,39]:

$${\sigma }_{\mathrm{dis}}=M_{\mathrm{T}}\alpha Gb\sqrt{\rho }$$

(4)

where α is a constant, and M, b, G, and ρ are constants (i.e., 0.166) for the average Taylor factor (2.75 for ferrite steel), the magnitude of the Burgers vector (0.248 nm for Fe), the shear modulus (78 GPa for Fe), and the average dislocation density, respectively [22,39]. ρ was determined experimentally from the XRD spectrum of each steel, as shown in Figure 9 and Table 2.

In micro-alloyed steels, as the precipitates, such as carbonitrides, observed in this work are too difficult to be cut by dislocations, precipitation strengthening is predominantly governed by the Orowan process, despite the fact that the cutting and Orowan mechanisms usually vie for prominence in materials containing precipitates [40,41]. Furthermore, the strength increase from the Orowan mechanism can be expressed as follows:

$${\sigma }_{\mathrm{ph}}={\displaystyle \frac{0.538Gb{f_{\mathrm{pre}}}^{\raisebox{1ex}{$1$}\!\left/ \!\raisebox{-1ex}{$2$}\right.}}{d_{\mathrm{pre}}}}\ln \left({\displaystyle \frac{d_{\mathrm{pre}}}{2b}}\right)$$

(5)

where σ_ph is the precipitation strengthening, G is the shear modulus (78 GPa), b is the Burgers vector equal to 0.248 nm, f_pre is the volume fraction of precipitates, and d_pre is the average diameter of the precipitates in as-tempered steel [40,41]. The solid-solution strengthening σ_ss can be obtained by subtracting the sum of σ₀ + σ_gb + σ_dis + σ_ph from the total yield strength of steel. The quantitative microstructural parameters used for strength modeling and all the individual strengthening contributions are summarized in Table 2 and

Table 4. Composition and corresponding increment of yield stress compared with the yield stress of Nb-V, V, and Nb steel.

Steel/MPa	σy/MPa	σ0/MPa	σss/MPa	σgb/MPa	σdis/MPa	σph/MPa
Nb-V	1077	40	233	161	300	343
V	1027	40	241	152	279	315
Nb	960	40	225	157	274	264

σ_y—yield strength; σ₀—lattice friction strengthening; σ_ss—solid-solution strengthening; σ_gb—grain refinement strengthening; σ_dis—dislocation strengthening; σ_ph—precipitation strengthening.

, respectively. It is found that more than 50% of strengthening contributions are from dislocation and precipitation.

Compared with V steel, the nano-sized secondary Nb precipitates formed during the quenching process in Nb-containing steel could effectively pin the grain boundaries, refining the austenite and substructures such as packets and blocks. Furthermore, when V was added in combination with Nb in Nb steel, the precipitation mass percentage increased, leading to an increase in the number of effective pinning particles, i.e., Nb-rich (Ti,Nb,V)(C,N) particles with diameters smaller than 50 nm, thereby enhancing the refinement effect. However, the simulation results showed that the difference in grain refinement strengthening among the three steels was not significant, indicating that the improvement in strengthening caused by this refinement was limited.

Compared with Nb steel, a large number of nano-sized MC phases precipitated in V steel during the tempering process. These precipitates are in a coherent or semi-coherent state with the matrix and provide strong precipitation strengthening through the Orowan mechanism. When Nb was added in combination with V in Nb-V steel, the composite addition of Nb could reduce the consumption of V during quenching, allowing more V to remain dissolved in the matrix and further enhance the precipitation strengthening by forming nano MC precipitates during tempering.

Additionally, at the same tempering temperature, the precipitates are the main factor affecting dislocation density. Nb-V steel maintains a higher number of dispersed precipitates in both the quenched and tempered states, resulting in a higher dislocation density. Nb steel and V steel maintain a higher number of dispersed precipitates only in the quenched and tempered states, respectively, leading to a relatively lower dislocation density and correspondingly lower dislocation strengthening.

5. Conclusions

The strengthening mechanism operating due to the addition of composite micro-alloying elements of Nb and V on the strengths and microstructure of steel has been investigated by comparing the benefits of micro-alloying itself related to the same basic composition. The following conclusions were obtained based on the systematic analysis:

• Nb-V composite micro-alloyed steel possessed the highest yield strength compared with Nb or V micro-alloyed steel when quenched at 870 °C and tempered at 450–650 °C. Furthermore, the strength increment of Nb-V micro-alloyed steel with respect to Nb or V micro-alloyed steel reached the maximum at a tempering temperature of 600 °C, and precipitation strengthening and dislocation strengthening presented higher strength contributions in Nb-V micro-alloyed steel than in Nb micro-alloyed steel and V micro-alloyed steel owing to the higher volume fraction and finer precipitate size.
• Compared with V steel, the nano-sized Nb-rich precipitates formed during the quenching process in Nb steel pinned the grain boundaries effectively and refined the austenite and substructures such as packets and blocks. When V was added in combination with Nb in steel, the precipitation temperature of the Nb-rich carbonitrides decreased and the mass percentage increased, which resulted in a higher volume fraction of effective pinning particles-Nb-rich (Ti,Nb,V)(C,N) with diameters smaller than 50 nm and led to an enhanced refinement of the prior austenite grain.
• Nb could reduce the consumption of V during quenching, allowing more V to be dissolved in the matrix after quenching, allowing more V to remain dissolved in the matrix, and further enhancing the precipitation strengthening by forming nano MC precipitates during tempering.
• Nb-V steel maintained a higher number of dispersed precipitates in both the quenched and tempered states, resulting in a higher dislocation density and dislocation strengthening.